blossom-v2-llama2-7B: benchmark results
A conversational instruction fine-tune released by Azure99 based on Llama-2-7b, using Azure99/blossom-chat-v1, Azure99/blossom-math-v1, ehartford/dolphin, WizardLM/WizardLM_evol_instruct_V2_196k. Training uses two stages; dataset example counts and epochs are not token or compute disclosures. Provider: Azure99. Access: Open.
Aggregate Rating 1314 ± 15, rank #8472 of 10916 rated models, from 68 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MMLU-by-task - High School Physics | 41.06 | Accuracy (%) | 92.3 |
| MMLU-by-task - College Mathematics | 38 | Accuracy (%) | 89.5 |
| MMLU-by-task - College Chemistry | 43 | Accuracy (%) | 83.5 |
| MMLU-by-task - Moral Scenarios | 40.34 | Accuracy (%) | 81.8 |
| MMLU-by-task - Professional Medicine | 55.88 | Accuracy (%) | 81.5 |
| MMLU-by-task - Formal Logic | 37.3 | Accuracy (%) | 79.6 |
| MMLU-by-task - High School European History | 69.09 | Accuracy (%) | 77.8 |
| MMLU-by-task - Security Studies | 63.27 | Accuracy (%) | 77.8 |
| MMLU-by-task - Business Ethics | 55 | Accuracy (%) | 74.7 |
| MMLU-by-task - High School World History | 73.84 | Accuracy (%) | 73.8 |
| MMLU-by-task - High School Statistics | 44.44 | Accuracy (%) | 71.5 |
| MMLU-by-task - Human Sexuality | 61.83 | Accuracy (%) | 71 |
Interactive version: theaggregate.ai/model?slug=blossom-v2-llama2-7b · How It Works · Data refreshed daily, snapshot 2026-11-08.